Solr的Scala客户端（scalikesolr）介绍

本文是scalikesolr的wiki的翻译

后边的代码片段使用了如下文档产生的索引"example/exampledocs/books.json".

{
  "id" : "978-0641723445",
  "cat" : ["book","hardcover"],
  "title" : "The Lightning Thief",
  "author" : "Rick Riordan",
  "series_t" : "Percy Jackson and the Olympians",
  "sequence_i" : 1,
  "genre_s" : "fantasy",
  "inStock" : true,
  "price" : 12.50,
  "pages_i" : 384
},
{
  "id" : "978-1423103349",
  "cat" : ["book","paperback"],
  "title" : "The Sea of Monsters",
  "author" : "Rick Riordan",
  "series_t" : "Percy Jackson and the Olympians",
  "sequence_i" : 2,
  "genre_s" : "fantasy",
  "inStock" : true,
  "price" : 6.49,
  "pages_i" : 304
}

查询

简单查询

使用核心查询参数和普通查询参数：

import com.github.seratch.scalikesolr._

val client = Solr.httpServer(new URL("http://localhost:8983/solr")).newClient
val request = new QueryRequest(writerType = WriterType.JavaBinary, query = Query("author:Rick")) // faster when using WriterType.JavaBinary
val response = client.doQuery(request)
println(response.responseHeader)
println(response.response)
response.response.documents foreach {
  case doc => {
    println(doc.get("id").toString()) // "978-1423103349"
    println(doc.get("cat").toListOrElse(Nil).toString) // List(book, hardcover)
    println(doc.get("title").toString()) // "The Sea of Monsters"
    println(doc.get("pages_i").toIntOrElse(0).toString) // 304
    println(doc.get("price").toDoubleOrElse(0.0).toString) // 6.49
  }
}

从SolrDocument绑定到对象

需要无参构造器和字段设置器。也可以指定有一个字符串作为参数的构造器的用户定义类型

case class PageI(val value: String = "")
case class Book(
  var id: String = "",
  var cat: List[String] = Nil,
  var price: Double = 0.0,
  var pageI: PageI = PageI(),
  var sequenceI: Int = 0 ) {
  def this() = {
    this ("", Nil, 0.0, PageI(), 0)
  }
}
val book = doc.bind(classOf[Book])
println(book.id) // "978-1423103349"
println(book.cat.size) // 2
println(book.price) // 6.49
println(book.pageI.value) // 304
println(book.sequenceI) // 2

使用高亮

使用高亮参数：

val request = new QueryRequest(
  writerType = WriterType.JSON, // but JSON format might be slow...
  query = Query("author:Rick"),
  sort = Sort("page_i desc")
)
request.highlighting = HighlightingParams(true)
val response = client.doQuery(request)
println(response.highlightings)
response.highlightings.keys foreach {
  case key => {
    println(key + " -> " + response.highlightings.get(key).get("author").toString)
    // "978-0641723445" -> "[i]Rick[/i] Riordan"
  }
}

使用MoreLikeThis

使用推荐：

val request = new QueryRequest(Query("author:Rick"))
request.moreLikeThis = MoreLikeThisParams(
  enabled = true,
  count = 3,
  fieldsToUseForSimilarity = FieldsToUseForSimilarity("body")
)
val response = client.doQuery(request)
println(response.moreLikeThis)
response.response.documents foreach {
  doc => {
    val id = doc.get("id").toString
    response.moreLikeThis.getList(id) foreach {
      case recommendation => {
        println(recommendation) // "SolrDocument(WriterType(standard),,Map(start -> 0, numFound -> 0))"
      }
    }
  }
}

使用多层面查询（FacetQuery）

使用简单的多层面查询参数:

val request = new QueryRequest(Query("author:Rick"))
request.facet = new FacetParams(
  enabled = true,
  params = List(new FacetParam(Param("facet.field"), Value("title")))
)
val response = client.doQuery(request)
println(response.facet.facetFields)
response.facet.facetFields.keys foreach {
  case key => {
    val facets = response.facet.facetFields.getOrElse(key, new SolrDocument())
    facets.keys foreach {
      case facetKey => println(facetKey + " -> " + facets.get(facetKey).toIntOrElse(0))
      // "thief" -> 1, "sea" -> 1, "monster" -> 1, "lightn" -> 1
    }
  }
}

使用结果集分组（Groupiong） /字段折叠（Field Collapsing）

val request = new QueryRequest(Query("genre_s:fantasy"))
request.group = new GroupParams(
  enabled = true,
  field = Field("author_t")
)
val response = client.doQuery(request)
println(response.groups.toString)
response.groups.groups foreach {
  case group => println(group.groupValue + " -> " + group.documents.toString)
  // "r.r" -> List(SolrDocument(...
  // "glen" -> List(SolrDocument(...
}

分布式查询

使用分布式查询：

val request = new QueryRequest(Query("genre_s:fantasy"))
request.shards = new DistributedSearchParams(
  shards = List(
    "localhost:8984/solr",
    "localhost:8985/solr"
  )
)
val response = client.doQuery(request)
println(response.groups.toString)

数据导入命令（DIH Command）

数据导入的命令:

val request = new DIHCommandRequest(command = "delta-import")
val response = client.doDIHCommand(request)
println(response.initArgs)
println(response.command)
println(response.status)
println(response.importResponse)
println(response.statusMessages)

文档更新

更新Solr索引的XML信息:

添加/更新文档

向Solr中添加文档:

val request = new UpdateRequest()
val doc1 = SolrDocument(
  writerType = WriterType.JSON,
  rawBody = """
  { "id" : "978-0641723445",
    "cat" : ["book","hardcover"],
    "title" : "The Lightning Thief",
    "author" : "Rick Riordan",
    "series_t" : "Percy Jackson and the Olympians",
    "sequence_i" : 1,
    "genre_s" : "fantasy",
    "inStock" : true,
    "price" : 12.50,
    "pages_i" : 384
  }"""
)
val doc2 = SolrDocument(
writerType = WriterType.JSON,
rawBody = """
  { "id" : "978-1423103349",
    "cat" : ["book","paperback"],
    "title" : "The Sea of Monsters",
    "author" : "Rick Riordan",
    "series_t" : "Percy Jackson and the Olympians",
    "sequence_i" : 2,
    "genre_s" : "fantasy",
    "inStock" : true,
    "price" : 6.49,
    "pages_i" : 304
  }"""
)
request.documents = List(doc1, doc2)
val response = client.doUpdateDocuments(request)
client.doCommit(new UpdateRequest)

删除文档

val request = new DeleteRequest(uniqueKeysToDelete = List("978-0641723445"))
val response = client.doDeleteDocuments(request)
client.doCommit(new UpdateRequest)
Commit

val response = client.doCommit(new UpdateRequest())
Rollback

val response = client.doRollback(new UpdateRequest())
Optimize

val response = client.doOptimize(new UpdateRequest())
Add / Update documents in CSV format

val request = new UpdateRequest(
  requestBody = "id,name,sequence_i\n0553573403,A Game of Thrones,1\n..."
)
val response = client.doUpdateDocumentsInCSV(request)
client.doCommit(new UpdateRequest)

以XML格式更新：

val request = new UpdateRequest(
  requestBody = "<optimize/>"
)
val response = client.doUpdateInXML(request)

以JSON格式更新：

val request = new UpdateRequest(
  writerType = WriterType.JSON,
  requestBody = "{ 7b1debea0390ffbb4e49e2f145a1625ea6ec7011quot;optimize7b1debea0390ffbb4e49e2f145a1625ea6ec7011quot;: { 7b1debea0390ffbb4e49e2f145a1625ea6ec7011quot;waitFlush7b1debea0390ffbb4e49e2f145a1625ea6ec7011quot;:false, 7b1debea0390ffbb4e49e2f145a1625ea6ec7011quot;waitSearcher7b1debea0390ffbb4e49e2f145a1625ea6ec7011quot;:false } }"
)
val response = client.doUpdateInJSON(request)

从更新格式中加载文档

不支持JSON格式

XML格式

val xmlString = "<add><doc><field name=ece0aa4e60e0be0d48ee70d07ac10cec4e21eecaquot;employeeIdece0aa4e60e0be0d48ee70d07ac10cec4e21eecaquot;>05991</field><field name=ece0aa4e60e0be0d48ee70d07ac10cec4e21eecaquot;officeece0aa4e60e0be0d48ee70d07ac10cec4e21eecaquot;>Bridgewater</field>..."
val docs = UpdateFormatLoader.fromXMLString(xmlString)
docs foreach {
  case doc => {
    println("employeeId:" + doc.get("employeeId").toString()) // "05991"
    println("office:" + doc.get("office").toString()) // "Bridgewater"
  }
}

CSV格式

val csvString = "id,name,sequence_i\n0553573403,A Game of Thrones,1\n..."
val docs = UpdateFormatLoader.fromCSVString(csvString)
docs foreach {
  case doc => {
    println(doc.get("id")) // "0553573403"
    println(doc.get("name")) // "A Game of Thrones"
    println(doc.get("sequence_i").toIntOrElse(0)) // 1
  }
}

Ping

val response = client.doPing(new PingRequest())
println(response.status) // "OK"